A study of urbanization analysis using remote sensing data

Urbanization analysis, especially urbanization level estimation has become an urgent task for city planning, urban resources management and risk assessment of urban disasters. DMSP/OLS stable lights imagery provides an accurate, economical, and straightforward way to analyze the urbanization pattern. Most previous studies had focused on patterns of urbanization development in provincial scale. In this study, we firstly extracted the night light compositive index(NLCI) from DMSP/OLS night light data. Based on it, we utilized NLCI to estimate the urbanization level of regions on the municipal scale. By regression analysis technique, the statistical model between NLCI and the urbanization level index(ULI) on the basis of socioeconomic data was established. And the ULIs of municipal regions in Mainland China were estimated by the model. Different from other studies, regionalization proposed by National Bureau of Statistics of China (NBSC) was introduced into the process of modeling in order to increase the accuracy of models. Results of modeling based on regionalization indicated that the regionalization had greatly strengthened the correlation between ULI and NLCI. Results of verification also demonstrated that there was strong consistency between estimated ULIs and actual statistical ULIs of many years, which showed the reliability and universality of the linear models.